Evolving ghost cytometry may take away labels for manufacturing of cell-based therapy products

Cell-based therapies, including stem cell therapies and regenerative medicine, offer transformative solutions for previously untreatable diseases. However, their manufacturing processes differ significantly from traditional drug production, requiring stringent Quality Control (QC) measures for the cells to ensure safety, efficacy and reproducibility. Fluorescence-based Flow Cytometry (FCM) is one of current QC methods, but rely on molecular markers that have potential impact on cell states and functionality, while also increasing costs. Moreover, there are cases that appropriate markers for accurately predicting cell function or differentiation are not available. Label-Free Ghost Cytometry (LF-GC) addresses these challenges by leveraging high-resolution and high-content morphological data and machine learning to classify and sort cells without fluorescent labels. LF-GC enables non-invasive, real-time analysis, preserving cell functionality and reducing manufacturing costs. Its applications extend beyond basic QC to include cell differentiation assessment and enrichment of potentially therapeutic cell populations. Recent studies have demonstrated its utility in analyzing blood and immune cells, induced pluripotent stem cells and retinal progenitor cells, highlighting its potential for improving cell manufacturing processes. Looking ahead, integrating LF-GC with unsupervised learning and other molecular techniques such as single cell sequencing will further expand the utility of LF-GC. As a scalable platform that can be automated, LF-GC has the potential to improve cell manufacturing by making advanced cell therapies safer, more accessible and cost-effective Keywords Cell-based therapy; Machine learning; Label-free cell analysis; Ghost cytometry